JP7571544B2 - HEALTH CONDITION DETERMINATION SYSTEM, HOUSE, MANAGEMENT DEVICE, PROGRAM, AND HEALTH CONDITION DETERMINATION METHOD - Google Patents

Description

The present invention relates to a health condition determination system, a house, a management device, a program, and a health condition determination method.

Patent Document 1 below discloses an abnormality determination device that includes a condition detection means that is attached to the subject's body and a determination means that is capable of wireless communication with the condition detection means. The condition detection means detects various conditions of the subject (body temperature, heart rate, posture, distance traveled, sleep state, etc.) using multiple sensors. The determination means determines whether or not an abnormality has occurred in the subject based on the results of condition detection by the condition detection means.

Patent No. 3378540

According to the abnormality determination device disclosed in the above-mentioned Patent Document 1, the condition detection means detects various conditions of the subject (body temperature, heart rate, posture, distance traveled, sleep state, etc.), but these conditions of the subject fluctuate greatly due to differences in the time period in which they are detected, or differences in the subject's daily activity schedule, etc. Therefore, even if the presence or absence of an abnormality in the subject is determined based on the detection results of conditions that fluctuate due to factors unrelated to such health conditions, the accuracy of the determination results is low.

In addition, because the condition detection means constantly detects various conditions of the subject and transmits them to the judgment means, the amount of data that the judgment means must process becomes enormous, increasing the processing load.

The present invention was made in consideration of these circumstances, and aims to provide a health condition determination system, house, management device, program, and health condition determination method that can improve the accuracy of determining a subject's health condition and reduce the processing load associated with the determination.

A health condition determination system according to one aspect of the present invention is a health condition determination system for determining the health condition of a subject residing in a house having stairs, and includes a first sensor that detects that the subject has woken up and outputs first data indicating the result of the detection, a second sensor that detects that the subject is located on one end step of the stairs and outputs second data indicating the result of the detection, a third sensor that detects that the subject is located on the other end step of the stairs and outputs third data indicating the result of the detection, and a management device capable of communicating with the first sensor, the second sensor, and the third sensor. The management device includes an acquisition unit that acquires the first data, the second data, and the third data, a calculation unit that calculates the time required for the subject to walk up the stairs from one end step to the other end step for the first time after getting up based on the first data, the second data, and the third data acquired by the acquisition unit, and outputs time data indicating the calculated time, a storage unit that stores a plurality of past time data, a determination unit that determines the subject's health condition on the day based on the time data on the day and the plurality of past time data, and an output unit that outputs the result of the determination made by the determination unit.

According to this aspect, the subject's health condition is determined based on the time it takes for the subject to walk from one end step to the other end step of the stairs for the first time after waking up. By measuring the first step after waking up, which is a daily behavior that is repeated at a roughly constant time every day and is generally mentally and physically unconsciously, factors that vary unrelated to the subject's health condition can be eliminated, and as a result, the accuracy of the health condition determination can be improved. In addition, since there are generally no branching paths or places to stop on the way up the stairs, by specifying the location for measuring the walking time as the stairs, factors in route selection that vary unrelated to the subject's health condition can be eliminated, and as a result, the accuracy of the health condition determination can be improved. Furthermore, by measuring a limited behavior in daily life, such as the first step up the stairs after waking up, the amount of data that the management device must process can be reduced, and as a result, the processing load associated with the health condition determination can be reduced.

In the above aspect, the device further includes a fourth sensor that detects the load applied to the handrail of the stairs by the subject and outputs fourth data indicating the result of the detection, the acquisition unit further acquires the fourth data, the calculation unit further calculates an average load applied to the handrail while the subject first walks up the stairs from the one end step to the other end step based on the fourth data acquired by the acquisition unit and outputs load data indicating the average load, the memory unit further stores a plurality of past load data, and the judgment unit judges the subject's health condition on the day based on the time data and load data of the day and the plurality of past time data and load data.

According to this aspect, the subject's health condition is assessed using not only the time it takes for the subject to walk up the stairs, but also the average load on the handrail at that time, which makes it possible to further improve the accuracy of the assessment.

In the above aspect, the calculation unit further calculates a first correlation graph showing the time series changes of the time data based on the past multiple time data, and the determination unit determines the subject's health condition on the day based on the time data on the day and the first correlation graph.

According to this aspect, the health condition is determined based on the first correlation graph that shows the time series changes in the time data, making it possible to easily and accurately determine the subject's health condition on that day.

In the above aspect, the calculation unit further calculates a slope value of the first correlation graph and outputs slope data indicating the slope value, the storage unit further stores a plurality of past slope data, the calculation unit further calculates a second correlation graph indicating a time series change in the slope data based on the plurality of past slope data, and the determination unit further determines the possibility of frailty of the subject based on the second correlation graph.

According to this aspect, it is possible to easily and accurately determine the possibility of a subject being frail based on the second correlation graph showing the time series changes in the slope data.

In the above aspect, the device further includes a group of sensors that detects movement of the subject from the subject's sleeping position to a position immediately before the one end step.

According to this aspect, the sensor group detects the movement of the subject from the sleeping position to the position just before one of the end steps, so it is possible to reliably identify that the measured stair walking is the subject's first stair walking after waking up.

A house according to one aspect of the present invention includes a staircase and a health condition determination system according to the above aspect.

A management device according to one aspect of the present invention is a management device provided with a health condition determination system for determining the health condition of a subject residing in a house having stairs, the health condition determination system including a first sensor that detects that the subject has woken up and outputs first data indicating the result of the detection, a second sensor that detects that the subject is located on one end step of the stairs and outputs second data indicating the result of the detection, a third sensor that detects that the subject is located on the other end step of the stairs and outputs third data indicating the result of the detection, and a management device capable of communicating with the first sensor, the second sensor, and the third sensor. The management device includes an acquisition unit that acquires the first data, the second data, and the third data, a calculation unit that calculates the time required for the subject to walk the stairs from one end step to the other end step for the first time after getting up based on the first data, the second data, and the third data acquired by the acquisition unit, and outputs time data indicating the calculated time, a storage unit that stores a plurality of past time data, a determination unit that determines the subject's health condition on the day based on the time data on the day and the plurality of past time data, and an output unit that outputs the result of the determination made by the determination unit.

A program according to one aspect of the present invention is a program for causing a computer to function as a management device provided in a health condition determination system for determining the health condition of a subject residing in a house having stairs, the health condition determination system including a first sensor that detects that the subject has woken up and outputs first data indicating the result of the detection, a second sensor that detects that the subject is located on one end step of the stairs and outputs second data indicating the result of the detection, a third sensor that detects that the subject is located on the other end step of the stairs and outputs third data indicating the result of the detection, and a control program for controlling the first sensor, the second sensor, and the health condition determination system. , and a management device capable of communicating with the third sensor, and the computer functions as: an acquisition means for acquiring the first data, the second data, and the third data; a calculation means for calculating the time required for the subject to walk the stairs from one end step to the other end step for the first time after getting up based on the first data, the second data, and the third data acquired by the acquisition means, and outputting time data indicating the calculated time; a determination means for determining the subject's health condition on the day based on the time data on the day and multiple pieces of time data from the past; and an output means for outputting the result of the determination made by the determination means.

A health condition determination method according to one aspect of the present invention is a health condition determination method for determining the health condition of a subject living in a house having stairs, in which a first sensor detects that the subject has woken up and outputs first data indicating the result of the detection, a second sensor detects that the subject is located at one end step of the stairs and outputs second data indicating the result of the detection, a third sensor detects that the subject is located at the other end step of the stairs and outputs third data indicating the result of the detection, and a management device acquires the first data, the second data, and the third data, calculates the time required for the subject to walk from the one end step to the other end step of the stairs for the first time after getting up based on the acquired first data, second data, and third data, outputs time data indicating the time, stores a plurality of past time data, and determines the subject's health condition for the day based on the time data for the day and the plurality of past time data, and outputs the result of the determination.

According to these aspects, the subject's health condition is determined based on the time it takes for the subject to walk up the stairs from one end step to the other end step for the first time after waking up. By measuring the first walk after waking up, which is a daily behavior that is repeated at a roughly constant time each day and is generally mentally and physically unconsciously, factors that vary unrelated to the subject's health condition can be eliminated, and as a result, the accuracy of the health condition determination can be improved. In addition, since there are generally no branching paths or places to stop on the way up the stairs, by specifying the location for measuring the walking time as the stairs, factors that vary unrelated to the subject's health condition can be eliminated, and as a result, the accuracy of the health condition determination can be improved. Furthermore, by measuring a limited behavior in daily life, such as the first walk up the stairs after waking up, the amount of data that the management device must process can be reduced, and as a result, the processing load associated with the health condition determination can be reduced.

The present invention makes it possible to improve the accuracy of determining a subject's health condition and reduce the processing load associated with the determination.

1 is a diagram showing a schematic configuration of a health condition determination system according to an embodiment of the present invention; FIG. 2 is a diagram showing a simplified configuration of a management device. 10 is a flowchart showing the process executed by a control unit regarding the generation of time data and load data. 11 is a flowchart showing a first example of processing executed by the control unit regarding determining a health condition of a subject. 13 is a flowchart showing a second example of the process executed by the control unit regarding determining the health condition of a subject. FIG. 13 is a diagram illustrating an example of a time correlation graph. FIG. 13 is a diagram showing an example of a slope correlation graph.

The following describes in detail the embodiments of the present invention with reference to the drawings. Note that elements with the same reference numerals in different drawings indicate the same or corresponding elements.

Figure 1 is a diagram showing a schematic configuration of a health condition determination system 100 according to an embodiment of the present invention. The health condition determination system 100 is constructed in a house H in which a user (hereinafter referred to as the "subject"), whose health condition is to be determined, resides.

In the example of Figure 1, house H is a two-story private home, with a corridor 8 and a living room (not shown) connected to it on the first floor, and bedroom 1, corridor 4, and landing 5 on the second floor. Bedroom 1 contains a bed 2 that the subject uses when sleeping. There is a door 3 between bedroom 1 and corridor 4. The end of corridor 4 as seen from bedroom 1 is landing 5. Corridor 8 and landing 5 are connected by a staircase 6 with a handrail 7. Note that bedroom 1 may be on the first floor and the living room on the second floor, or bedroom 1 may be in the basement and the living room on the first floor.

The health condition determination system 100 includes a wake-up sensor 12, a bed exit sensor 13, an opening/closing sensor 14, human presence sensors 15 and 16, load sensors 17 to 19, and a management device 11. The wake-up sensor 12, bed exit sensor 13, opening/closing sensor 14, human presence sensors 15 and 16, and load sensors 17 to 19 can communicate with the management device 11 via any communication method such as Wi-Fi.

The wake-up sensor 12 is, for example, a sleep meter that can be worn on the subject's body (e.g., a wristwatch), a sleep meter embedded in the mattress of the bed 2, or a sleep meter installed on the ceiling surface of the bedroom 1. The wake-up sensor 12 (first sensor) detects that the subject has woken up by detecting that the subject has woken up from a sleep state of at least a certain period of time (e.g., three hours). When the wake-up sensor 12 detects that the subject has woken up, it transmits wake-up detection data indicating the result of the detection to the management device 11.

The bed exit sensor 13 is, for example, a camera that captures images of the area around the bed 2, or a load sensor embedded in the bed 2. The bed exit sensor 13 detects that the subject has left the bed 2. When the bed exit sensor 13 detects that the subject has left the bed, it transmits bed exit detection data indicating the detection result to the management device 11.

The opening/closing sensor 14 is, for example, a magnetic sensor embedded in the door 3. The opening/closing sensor 14 detects when the door 3 is opened and when the door 3 is closed. When the opening/closing sensor 14 detects whether the door 3 is open or closed, it transmits opening/closing detection data indicating the result of the detection to the management device 11.

The human presence sensor 15 is installed on the opposite side of the door 3 as viewed from the bedroom 1. The human presence sensor 15 is, for example, an infrared sensor installed on the ceiling or a load sensor embedded in the floor. The human presence sensor 15 detects that the subject has left the bedroom 1. When the human presence sensor 15 detects that the subject has left the bedroom 1, it transmits exit detection data indicating the detection result to the management device 11.

The human presence sensor 16 is installed on the landing 5. The human presence sensor 16 is, for example, an infrared sensor installed on the ceiling or a load sensor embedded in the floor. The human presence sensor 16 detects that the subject is located on the landing 5 (i.e., the position immediately before the top step of the stairs 6). When the human presence sensor 16 detects that the subject is located on the landing 5, it transmits position detection data indicating the detection result to the management device 11.

The load sensor 17 is embedded in the top step of the stairs 6. The load sensor 17 (second sensor) detects that the subject is located at the top step of the stairs 6. When the load sensor 17 detects that the subject is located at the top step of the stairs 6, it transmits position detection data indicating the detection result to the management device 11.

The load sensor 18 is embedded in the bottom step of the stairs 6. The load sensor 18 (third sensor) detects that the subject is located at the bottom step of the stairs 6. When the load sensor 18 detects that the subject is located at the bottom step of the stairs 6, it transmits position detection data indicating the detection result to the management device 11.

The load sensor 19 is embedded in a support member for attaching the handrail 7 to a wall. If there are multiple support members, the load sensor 19 is embedded in each of the multiple support members. The load sensor 19 (fourth sensor) detects the load applied to the handrail 7 by the subject. When the load sensor 19 detects the load applied to the handrail 7 by the subject, it transmits load value data indicating the detected load value to the management device 11.

The management device 11 can communicate with the display device 22 by any communication method such as Wi-Fi. The display device 22 is, for example, a smartphone used by the subject.

The management device 11 can communicate with a server device 21, such as a cloud server, via any communication network 20, such as a public line network, using any communication method, such as IP.

Figure 2 is a simplified diagram showing the configuration of the management device 11. The management device 11 includes a control unit 31, a memory unit 32, and communication units 33 and 34.

The control unit 31 is configured using a data processing device such as a CPU. The memory unit 32 is configured using a non-volatile memory device such as a hard disk or semiconductor memory. The communication unit 33 is configured using a communication module compatible with any communication method such as Wi-Fi for communicating with the wake-up sensor 12, bed exit sensor 13, open/close sensor 14, human presence sensors 15, 16, load sensors 17-19, and display device 22. The communication unit 34 is configured using a communication module compatible with any communication method such as IP for communicating with the server device 21.

The memory unit 32 stores a program 51, time data 52, load data 53, and tilt data 54.

When the CPU executes the program read from the memory unit 32, the control unit 31 functions as an acquisition unit 41, a calculation unit 42, a determination unit 43, and an output unit 44. In other words, the program 51 is a program for causing the computer serving as the management device 11 to function as the acquisition unit 41, the calculation unit 42, the determination unit 43, and the output unit 44.

Figure 3 is a flowchart showing the processing executed by the control unit 31 regarding the generation of time data 52 and load data 53.

First, in step SP101, when the subject wakes up and the wake-up sensor 12 transmits wake-up detection data, the communication unit 33 receives the wake-up detection data, and the acquisition unit 41 acquires the wake-up detection data from the communication unit 33. The process of step SP101 is repeatedly executed until the acquisition unit 41 acquires the wake-up detection data (i.e., until the subject wakes up).

Next, in step SP102, when the subject gets out of bed and the bed exit sensor 13 transmits the bed exit detection data, the communication unit 33 receives the bed exit detection data, and the acquisition unit 41 acquires the bed exit detection data from the communication unit 33. The process of step SP102 is repeatedly executed until the acquisition unit 41 acquires the bed exit detection data (i.e., until the subject gets out of bed). Note that if the subject starts sleeping again without getting out of bed, the process returns to step SP101.

Next, in step SP103, when the subject opens and closes the door 3 and the opening and closing sensor 14 transmits the opening and closing detection data, the communication unit 33 receives the opening and closing detection data, and the acquisition unit 41 acquires the opening and closing detection data from the communication unit 33. The process of step SP103 is repeatedly executed until the acquisition unit 41 acquires the opening and closing detection data (i.e., until the subject opens and closes the door 3). Note that if the subject gets back into bed 2 without opening or closing the door 3, the process returns to step SP102.

Next, in step SP104, when the subject leaves bedroom 1 and the human presence sensor 15 transmits exit detection data, the communication unit 33 receives the exit detection data, and the acquisition unit 41 acquires the exit detection data from the communication unit 33. The process of step SP104 is repeatedly executed until the acquisition unit 41 acquires the exit detection data (i.e., until the subject leaves bedroom 1).

Next, in step SP105, when the target person reaches the landing 5 and the human presence sensor 16 transmits position detection data, the communication unit 33 receives the position detection data, and the acquisition unit 41 acquires the position detection data from the communication unit 33. The process of step SP105 is repeatedly executed until the acquisition unit 41 acquires the position detection data (i.e., until the target person reaches the landing 5).

Next, in step SP106, when the subject reaches the top step of the stairs 6 and the load sensor 17 transmits position detection data, the communication unit 33 receives the position detection data, and the acquisition unit 41 acquires the position detection data from the communication unit 33. The process of step SP106 is repeatedly executed until the acquisition unit 41 acquires the position detection data (i.e., until the subject reaches the top step of the stairs 6).

When the subject reaches the top step of the stairs 6, in step SP107, the control unit 31 causes the load sensor 19 to start measuring the load by sending a control signal from the communication unit 33 to the load sensor 19. When the load sensor 19 sends the load value data, the communication unit 33 receives the load value data, and the acquisition unit 41 acquires the load value data from the communication unit 33.

Next, in step SP108, when the subject reaches the bottom step of the stairs 6 and the load sensor 17 transmits position detection data, the communication unit 33 receives the position detection data, and the acquisition unit 41 acquires the position detection data from the communication unit 33. The process of step SP108 is repeatedly executed until the acquisition unit 41 acquires the position detection data (i.e., until the subject reaches the bottom step of the stairs 6).

When the subject reaches the bottom step of the stairs 6, in step SP109, the control unit 31 causes the load sensor 19 to end the load measurement by sending a control signal from the communication unit 33 to the load sensor 19.

Next, in step SP110, the calculation unit 42 calculates the time required for the subject to walk from the top step to the bottom step of the stairs 6 (required time T0 for the day) by calculating the time difference between the time when the acquisition unit 41 acquires the position detection data transmitted from the load sensor 17 and the time when the acquisition unit 41 acquires the position detection data transmitted from the load sensor 18. The control unit 31 generates time data 52 including the required time and date information, and stores the time data 52 in the memory unit 32. The memory unit 32 also accumulates and stores time data 52 related to multiple required times in the past up to yesterday. The calculation unit 42 also calculates the integrated value of the load value data acquired by the acquisition unit 41 during the required time, and divides the integrated value by the required time to calculate the average load value (average load value L0 for the day) on the handrail 7 while the subject walks from the top step to the bottom step of the stairs 6. The control unit 31 generates load data 53 including the weighted average value and date information, and stores the load data 53 in the memory unit 32. The memory unit 32 also accumulates and stores load data 53 related to multiple weighted average values from the past up until yesterday.

Figure 4 is a flowchart showing a first example of the processing executed by the control unit 31 to determine the subject's health condition.

First, in step SP201, the calculation unit 42 reads out from the storage unit 32 a number of time data 52 relating to a predetermined period of time up to the previous day (e.g., the most recent week or month), and calculates the average value of the multiple required times indicated by the multiple time data 52 to calculate a reference required time T1.

Next, in step SP202, the determination unit 43 reads the time data 52 for the current day from the storage unit 32, and determines whether the required time T0 for the current day indicated by the time data 52 for the current day is less than the reference required time T1.

If the time required for that day T0 is less than the reference time required for that day T1 (step SP202: YES), then in step SP203, the judgment unit 43 judges that the subject's health condition is good C1. In this case, the output unit 44 outputs result data representing the condition C1 as the result of the judgment by the judgment unit 43, and the communication unit 33 transmits the result data to the display device 22. The display device 22, which has received the result data representing the condition C1, displays on the screen a graphic or text message indicating that the health condition for that day is good. Note that the manner in which the judgment result is notified to the user may be audio output from a speaker, rather than a display on the screen (same below).

If the required time T0 for the day is equal to or greater than the reference required time T1 (step SP202: NO), then in step SP204, the calculation unit 42 reads from the memory unit 32 a number of load data 53 relating to a specified period up to the previous day (e.g., the most recent week to month), and calculates the average value of the multiple load average values indicated by the plurality of load data 53 to calculate the reference load average value L1.

Next, in step SP205, the determination unit 43 reads the load data 53 for the current day from the storage unit 32, and determines whether the average load value L0 for the current day indicated by the load data 53 for the current day is equal to or less than the reference average load value L1.

If the load average value L0 for that day is equal to or less than the reference load average value L1 (step SP205: YES), then in step SP206, the judgment unit 43 judges that the subject's health condition is a slightly poor condition C2. In this case, the output unit 44 outputs result data representing the condition C2 as the result of the judgment by the judgment unit 43, and the communication unit 33 transmits the result data to the display device 22. Having received the result data representing the condition C2, the display device 22 displays on the screen a graphic or text message indicating that the health condition for that day is slightly poor.

If the load average value L0 for that day exceeds the reference load average value L1 (step SP205: NO), then in step SP207, the judgment unit 43 judges that the subject's health condition is poor condition C3. In this case, the output unit 44 outputs result data representing condition C3 as the result of the judgment by the judgment unit 43, and the communication unit 33 transmits the result data to the display device 22. Having received the result data representing condition C3, the display device 22 displays on the screen a graphic or text message indicating that the health condition for that day is poor.

Figure 5 is a flowchart showing a second example of the processing executed by the control unit 31 to determine the subject's health condition.

First, in step SP301, the calculation unit 42 reads out from the storage unit 32 a plurality of time data 52 relating to a specified period (e.g., 30 days) including the current day, and applies any algorithm for deriving an approximate straight line, such as the least squares method, to the multiple required times indicated by the multiple time data 52. As a result, the calculation unit 42 calculates a time correlation graph G1 showing the time series change in the required time based on the multiple time data 52 for the current day and the past. FIG. 6 is a diagram showing an example of a time correlation graph G1 calculated for the required period for the most recent 30 days. The calculation unit 42 generates slope data 54 including the slope value of the straight line of the time correlation graph G1 and date information, and stores it in the storage unit 32 as the slope data 54 for the current day. The storage unit 32 also accumulates and stores slope data 54 relating to multiple past slope values up to yesterday.

Next, in step SP302, the determination unit 43 determines whether the required time T0 for that day, which is indicated by the time data 52 for that day, is equal to or less than the estimated value for that day, which is indicated as a value on the time correlation graph G1.

If the time required for that day T0 is equal to or less than the estimated value for that day (step SP302: YES), then in step SP303, the determination unit 43 determines that the subject's health condition is good F1. In this case, the output unit 44 outputs result data representing the condition F1 as the result of the determination by the determination unit 43, and the communication unit 33 transmits the result data to the display device 22. Having received the result data representing the condition F1, the display device 22 displays on the screen a graphic or text message indicating that the health condition for that day is good.

If the required time T0 for the day exceeds the estimated value for the day (step SP302: NO), then in step SP304, the calculation unit 42 reads out from the storage unit 32 a plurality of slope data 54 relating to a recent predetermined period (e.g., one year) including the current day, and applies any approximate curve derivation algorithm such as the cosinor method to the plurality of slope values indicated by the plurality of slope data 54. As a result, the calculation unit 42 calculates a slope correlation graph G2 indicating a time series change in the slope value based on the plurality of slope data 54 for the current day and the past. FIG. 7 is a diagram showing an example of a slope correlation graph G2 calculated for the slope values for the past year. As shown in the example of FIG. 7, the calculation unit 42 may calculate an average value for each month (or week) by averaging the slope values for the past year by month (or week), and calculate the slope correlation graph G2 based on the plurality of average values.

Next, in step SP305, the determination unit 43 determines whether the slope correlation graph G2 has an upward trend (i.e., the older the data, the smaller the slope value, and the newer the data, the larger the slope value).

If the slope correlation graph G2 does not have an upward trend to the right (step SP305: NO), for example, if the slope correlation graph G2 has a downward trend or periodicity, then in step SP306, the judgment unit 43 judges that the subject is in state F2, which indicates a low possibility of frailty (a state between a healthy state and a state requiring care). In this case, the output unit 44 outputs result data indicating state F2 as the result of the judgment by the judgment unit 43, and the communication unit 33 transmits the result data to the display device 22. Having received the result data indicating state F2, the display device 22 displays on the screen a graphic or text message indicating that the possibility of frailty is low.

If the slope correlation graph G2 has an upward trend to the right (step SP305: YES), then in step SP307, the determination unit 43 determines that the subject is in state F3, which indicates a high possibility of frailty. In this case, the output unit 44 outputs result data indicating state F3 as the result of the determination by the determination unit 43, and the communication unit 33 transmits the result data to the display device 22. Having received the result data indicating state F3, the display device 22 displays on the screen a graphic or text message indicating a high possibility of frailty and a recommendation to visit a medical institution.

In the above description, the judgment unit 43 judges the possibility of frailty based on whether the slope correlation graph G2 has an upward trend to the right, but other judgment algorithms may be used. For example, the server device 21 collects slope data 54 related to a large number of users from the management device 11 of each user via the communication network 20. The large number of users includes various users with different attributes such as age, sex, occupation, and place of residence, and also includes both frail and non-frail users. The server device 21 also collects this attribute information and frailty information from the management device 11 of each user via the communication network 20. The server device 21 uses this information collected from the management devices 11 of the large number of users to create a judgment model for judging the possibility of frailty by deep learning such as a neural network. The server device 21 transmits the created judgment model to the management device 11 via the communication network 20. The management device 11 uses the judgment model to judge the possibility of frailty based on the attribute information and slope data 54 of the subject.

In this way, according to the health condition determination system 100 of the present embodiment, the health condition of the subject is determined based on the time required for the subject to walk from the top step (one end step) to the bottom step (the other end step) of the stairs 6 for the first time after waking up. By measuring the first walk after waking up, which is a daily behavior that is repeated every day at a generally fixed time and mentally and physically almost unconsciously, factors that vary unrelated to the subject's health condition can be eliminated, and as a result, the accuracy of the health condition determination can be improved. In addition, since there are generally no branching paths or stopovers on the stairs 6, by specifying the location for measuring the walking time as the stairs 6, factors of route selection that vary unrelated to the subject's health condition can be eliminated, and as a result, the accuracy of the health condition determination can be improved. Furthermore, by measuring a limited behavior in daily life, such as the first walk up the stairs after waking up, the amount of data to be processed by the management device 11 can be reduced, and as a result, the processing load associated with the health condition determination can be reduced.

In addition, according to the health condition determination system 100 of this embodiment, the subject's health condition is determined not only based on the time it takes for the subject to walk up the stairs 6, but also based on the average load on the handrail 7 at that time, which makes it possible to further improve the accuracy of the determination.

In addition, according to the health condition determination system 100 of this embodiment, the health condition is determined based on the time correlation graph G1 (first correlation graph) that shows the time series changes in the time data 52, making it possible to easily and accurately determine the subject's health condition on the day.

In addition, according to the health condition determination system 100 of this embodiment, it is possible to easily and accurately determine the possibility of frailty in a subject based on the slope correlation graph G2 (second correlation graph) that shows the time series changes in the slope data 54.

In addition, according to the health condition determination system 100 of this embodiment, the movement of the subject from the bed 2 to the landing 5 is detected by a group of sensors including the bed exit sensor 13, the opening/closing sensor 14, and the human presence sensors 15 and 16. Therefore, it is possible to reliably identify that the measured stair walking is the subject's first stair walking after getting up.

6 Stairs 11 Management device 12 Wake-up sensor 13 Bed exit sensor 14 Open/close sensor 15, 16 Human presence sensor 17-19 Load sensor 31 Control unit 32 Memory unit 41 Acquisition unit 42 Calculation unit 43 Determination unit 44 Output unit 51 Program 52 Time data 53 Load data 54 Tilt data 100 Health condition determination system G1 Time correlation graph G2 Tilt correlation graph H House

Claims (9)

A health condition determination system for determining a health condition of a subject residing in a house having stairs, comprising:
A first sensor that detects that the subject has woken up and outputs first data indicating a result of the detection;
a second sensor that detects that the subject is located on one end step of the staircase and outputs second data indicating the result of the detection;
a third sensor that detects that the subject is located at the other end step of the staircase and outputs third data indicating the result of the detection;
a management device capable of communicating with the first sensor, the second sensor, and the third sensor;
Equipped with
The management device includes:
an acquisition unit that acquires the first data, the second data, and the third data;
a calculation unit that calculates a time required for the subject to walk up the stairs from the one end step to the other end step for the first time after getting up, based on the first data, the second data, and the third data acquired by the acquisition unit, and outputs time data indicating the calculated time;
A storage unit that stores a plurality of past time data;
A determination unit that determines a health condition of the subject on the day based on the time data on the day and a plurality of past time data;
an output unit that outputs a result of the determination by the determination unit;
A health condition determination system having the above structure. A fourth sensor is further provided to detect a load applied by the subject to the handrail of the staircase and output fourth data indicating the result of the detection,
The acquisition unit further acquires the fourth data,
the calculation unit further calculates a load value on the handrail while the subject walks up the stairs for the first time from the one end step to the other end step after getting up, based on the fourth data acquired by the acquisition unit, and outputs load data indicating the load value;
The storage unit further stores a plurality of past load data,
The health condition determination system according to claim 1 , wherein the determination unit determines the subject's health condition on the day based on the time data and the load data on the day and a plurality of the time data and a plurality of the load data in the past. The calculation unit further calculates a first correlation graph indicating a time series change of the time data based on a plurality of past time data;
The health condition determination system according to claim 1 , wherein the determination unit determines the subject's health condition on that day based on the time data on that day and the first correlation graph. The calculation unit further calculates a slope value of the first correlation graph and outputs slope data indicating the slope value;
The storage unit further stores a plurality of past tilt data,
The calculation unit further calculates a second correlation graph indicating a time series change of the tilt data based on a plurality of past tilt data;
The health condition determination system according to claim 3 , wherein the determination unit further determines a possibility of frailty of the subject based on the second correlation graph. The health condition determination system according to any one of claims 1 to 4, further comprising a group of sensors that detects the movement of the subject from the subject's sleeping position to a position immediately before the one end step. Stairs and
A health condition determination system according to any one of claims 1 to 5,
A house equipped with: A management device provided in a health condition determination system for determining a health condition of a subject residing in a house having stairs,
The health condition determination system includes:
A first sensor that detects that the subject has woken up and outputs first data indicating a result of the detection;
a second sensor that detects that the subject is located on one end step of the staircase and outputs second data indicating the result of the detection;
a third sensor that detects that the subject is located at the other end step of the staircase and outputs third data indicating the result of the detection;
a management device capable of communicating with the first sensor, the second sensor, and the third sensor;
Equipped with
The management device includes:
an acquisition unit that acquires the first data, the second data, and the third data;
a calculation unit that calculates a time required for the subject to walk up the stairs from the one end step to the other end step for the first time after getting up, based on the first data, the second data, and the third data acquired by the acquisition unit, and outputs time data indicating the calculated time;
A storage unit that stores a plurality of past time data;
A determination unit that determines a health condition of the subject on the day based on the time data on the day and a plurality of past time data;
an output unit that outputs a result of the determination by the determination unit;
A management device comprising: A program for causing a computer to function as a management device included in a health condition determination system for determining the health condition of a subject living in a house having stairs, comprising:
The health condition determination system includes:
A first sensor that detects that the subject has woken up and outputs first data indicating a result of the detection;
a second sensor that detects that the subject is located on one end step of the staircase and outputs second data indicating the result of the detection;
a third sensor that detects that the subject is located at the other end step of the staircase and outputs third data indicating the result of the detection;
a management device capable of communicating with the first sensor, the second sensor, and the third sensor;
Equipped with
The computer,
an acquisition means for acquiring the first data, the second data, and the third data;
a calculation means for calculating a time required for the subject to walk up the stairs from the one end step to the other end step for the first time after getting up, based on the first data, the second data, and the third data acquired by the acquisition means, and outputting time data indicating the calculated time;
A determination means for determining the subject's health condition on the day based on the time data on the day and a plurality of past time data;
an output means for outputting a result of the determination by the determination means;
A program to function as a A health condition determination method for determining the health condition of a subject residing in a house having stairs, comprising:
A first sensor detects that the subject has woken up and outputs first data indicating a result of the detection;
a second sensor detects that the subject is located on one end step of the staircase and outputs second data indicating the result of the detection;
a third sensor detects that the subject is located at the other end step of the staircase and outputs third data indicating the result of the detection;
The management device:
acquiring the first data, the second data, and the third data;
calculating a time required for the subject to walk up the stairs from the one end step to the other end step for the first time after getting up based on the acquired first data, the second data, and the third data, and outputting time data indicating the calculated time;
storing a plurality of past time data;
determining a health condition of the subject on that day based on the time data on that day and a plurality of past time data;
The health condition determination method outputs the result of the determination.